culture–grown plants. We do not know whether this regulation occurs specifically at the level of rbcL
mRNA translation initiation.
To gain insight into whether rbcL mRNA expression is limited by rbcL DNA template availability,
Jiang and Rodermel [13] performed genomic Southern blot analyses. In these experiments, equal amounts
of total cell DNA from the wild-type and mutant leaves were blotted onto filters and the filters were
probed with rbcL DNA sequences. These analyses showed that rbcL DNA levels fall in concert with rbcL
mRNA as one moves down the canopy. The patterns of change (and the magnitude of the change) were
similar in both sets of plants. Although we do not know whether each of the polyploid chloroplast DNAs
is equally capable of being transcribed in leaves from these plants, these data suggest that developmental
controls on rbcL mRNA abundance may be exerted, at least in part, by rbcL template availability in both
wild-type and Rubisco antisense tobacco plants.
The decreases in Rubisco appear to have little impact on the accumulation of proteins other than Ru-
bisco [13]. This indicates that leaf developmental programming is generally insensitive to Rubisco con-
centrations. This conclusion is consistent with the observations of others who have examined protein ac-
cumulation in first fully expanded leaves of antisense plants with up to ~80% reductions in Rubisco
content [46,49,50]. The protein accumulation profiles to date have relied on Western immunoblot analy-
ses of relatively few “representative” proteins; it is now possible to conduct detailed proteomics analyses
to confirm this generalization.
C. Development of an Individual Leaf
Miller et al. [27] have recently studied the development of individual antisense and wild-type leaves. In
these experiments, leaves were sampled throughout their ontogeny (similar to the elevated CO 2 studies).
“Developmentally similar” leaves were used. These were leaves from node 13 in the antisense plants and
from node 10 in the wild-type plants. Both of these leaves emerge during the fast-growth phase of shoot
morphogenesis and have similar characteristics (final size, canopy position, photosynthetic rate) [41]. Be-
cause many of the analyses required destructive sampling, developmentally similar leaves were isolated
from many plants. “Day 1” status was accorded to the leaves when they first attained a size sufficient for
analysis (~3 cm in width and ~5 cm in length).
We first examined various photosynthetic parameters in the mutant and wild-type leaves. As illus-
trated in Figure 1, photosynthetic rates increased to a maximum on day 12 in the wild-type plants, then
declined steadily until they fell below zero on day 35; shortly thereafter they abscised from the plant. Al-
though maximal rates were somewhat lower in the antisense leaves, they were relatively constant until
about day 20, after which they declined steadily until day 30. Thereafter, they remained fairly constant
and did not fall below zero, even at day 55. Antisense leaves did not abscise until around day 60. Similar
patterns of change were observed for chlorophyll concentrations, Rubisco contents, and Rubisco activi-
ties. Taken together, these data indicate that the antisense leaves are longer lived than wild-type leaves
and that this increase in longevity is due to a prolongation of the senescence phase of development.
Much of this prolongation appears to be due to alterations in the expression of genes for photosyn-
thetic proteins. During wild-type leaf development, we found that the senescence phase is marked by a
progressive decline in the content of total cell protein, chloroplast rRNA, and chloroplast DNA. These pa-
rameters followed similar patterns of change in the antisense leaves, with the exception that the senescent
declines were markedly prolonged. mRNAs for specific photosynthetic proteins also decreased during the
senescence of wild-type and antisense leaves. For instance, rbcS and rbcL mRNAs decreased in parallel
in the wild type, consistent with the hypothesis that coordinate changes in rbcS and rbcL mRNA abun-
dance play a central role in determining Rubisco content during wild-type leaf ontogeny. As discussed
earlier, this coordination has been observed in all of our developmental studies on wild-type tobacco
[13,28,40]. By contrast, rbcS and rbcL mRNAs undergo a longer senescent decline in the antisense
leaves. Also as observed in our earlier studies, the content of the holoenzyme in the antisense leaves ap-
pears to be regulated primarily at the level of rbcS transcript accumulation; i.e., LS protein accumulation
is regulated posttranscriptionally.
V. CONCLUSIONS
Examination of the Rubisco antisense mutants has revealed that decreased source strength regulates the
duration and progression of tobacco leaf senescence. Increased source strength (elevated CO 2 ), on the
SOURCE STRENGTH AND LEAF DEVELOPMENT 123